Background and objectives: Although research suggests that introducing varying levels of fear during exposure enhances outcomes for some anxiety-related problems, this has not been examined in the context of obsessions. The current preliminary study tested the hypothesis that introducing variability in exposure intensity would improve long-term outcomes relative to traditional gradual (hierarchical) exposure Methods: Adults (N = 40) with a moderately distressing unacceptable obsessional thought were randomly assigned in parallel to four twice-weekly sessions of: (a) gradual exposure (EXP-G; n = 19) emphasizing hierarchical exposure completion, or (b) variable exposure (EXP-V; n = 21) emphasizing variability in exposure intensity Results: There were no significant differences in pre to post changes between groups using self-report, interview, or behavioral outcomes (as evaluated by an independent assessor blind to treatment condition). Group comparisons at 3-month follow-up did not reach statistical significance but were moderate in magnitude. Specifically, as measured by clinical interview (the Yale-Brown Obsessive-Compulsive Scale; primary outcome) and self-report, individuals in the EXP-G group maintained gains at 3-month follow-up, while the EXP-V group continued to improve. Treatment expectancies and satisfaction were comparable for both groups. Five participants withdrew from the EXP-G condition, and none withdrew from the EXP-V condition. In contrast to previous studies, variability in subjective and physiological fear during exposure did not predict outcomes Limitations: The study employed an analogue sample with moderate unacceptable obsessions, and results should be replicated in clinical samples Conclusions: Variable exposure warrants future study to understand the mechanisms, moderators, and implications of this novel approach